ESG Rating Disagreement and Stock Returns

ESG Rating Disagreement and Stock Returns

Rajna Gibson Brandon, Philipp Krueger, Nadine Riand, Peter Steffen Schmidt

March 02 2020

A variety of professional data vendors nowadays provide firm-level Environmental, Social, and Governance (ESG) ratings (or scores). These ratings increasingly shape investment decisions of institutional investors. Since these ESG ratings are often based on non-standardized information and data vendors use different rating methodologies, ratings can diverge substantially. So far, it is not well understood why ESG ratings diverge and whether, or not, there are financial implications from such divergence. In this paper, we aim to fill this gap by (i) quantifying the magnitude of disagreement between ESG ratings from six well-known ESG ratings providers, (ii) identifying whether disagreement is higher for certain types of firms, and (iii) examining whether ESG rating disagreement affects future stock returns.

The first two issues are rather descriptive in nature, and therefore we do not have any specific ex-ante expectations about neither the magnitude of disagreement nor the question of whether it is higher or lower for certain types of firms. In contrast, regarding the relation between stock returns and ESG rating disagreement, we postulate two competing hypotheses: A risk-based explanation, under which higher ESG rating disagreement results in higher future returns because investors need to be compensated for a higher level of risk. Under the competing optimism-bias based explanation, higher disagreement results in lower future returns because investors have been too optimistic about high-disagreement stocks. We test these hypotheses for the overall ESG rating and separately for the three E, S, and G components. Given that recent research suggests that ESG ratings and a country’s legal origin are related, we ask whether this idea carries over to the legal origin of the ESG rating providers themselves. We assign rating providers into two groups: First, rating providers with a civil law background and secondly rating providers with a common law background. Based on the idea that firm governance in civil law countries is more stakeholder oriented, we conjecture that for the optimism bias hypothesis, disagreement among civil law raters should be more important for social ratings. In contrast, disagreement among common law rating about the governance rating should be more informative, mainly because governance in common law countries is more shareholder-centric.

We conduct our empirical analysis using a sample of S&P 500 firms spanning from 2013 to 2017. We use ESG ratings from six prominent ESG rating providers, namely Asset4 (now Refinitiv ESG), Sustainalytics, Inrate, Bloomberg, MSCI KLD, and MSCI IVA.

We report substantial disagreement across ESG ratings providers. While the average correlation for the overall ESG and the Environmental rating across the providers is 0.46 and 0.43, we report smaller numbers, that is more disagreement, for the Social and Governance ratings, namely 0.33 and 0.19. There are several firm characteristics that are related to rating disagreement, most importantly gross profitability, a missing credit rating, and market capitalization. Firms with higher gross profitability have lower rating disagreement, whereas firms without a credit rating and a higher market capitalization are subject to higher ESG rating disagreement. Finally, we find that ESG ratings disagreement affects subsequent stock returns in an intricate way. Namely, we observe that for the dispersion of the environmental rating the risk-based explanation is validated, whereas for disagreement about the social and governance ratings, the optimism-bias based explanation seems to be at play for civil and common law based data providers.

To the best of our knowledge, our paper is the first to document the multi-faceted implications of ESG ratings disagreement on stock returns and thus on firms’ costs of capital. Our analysis also has important implications for responsible investors who rely on one or a couple of ESG rating sources and ignore the impact of ESG ratings disagreement on future stock prices and thus do not factor in the potential pricing implications of ESG rating disagreement on the performance of responsible investment strategies.

 

Authors

Real name:
Nadine Riand
Real name:
Peter Steffen Schmidt